Glycolytic capacity instructs tumour vasculature and response to immunotherapy

Keywords

• Tumour angiogenesis

• Glycolysis

• Immune cell trafficking

Main Findings

Immune checkpoint inhibitors have transformed the care of cancer patients, but efficacy often remains limited to only a small subset of patients. Tumour metabolic adaptation is increasingly recognized as a barrier to immunotherapy efficacy. Aerobic glycolysis, or the Warburgh effect, is a hallmark of cancer cells that not only fuels tumour growth, but also promotes angiogenesis, generating disorganized, leaky vasculature that impairs T cell infiltration into tumours. While the links between tumour glycolysis, hypoxia and angiogenesis are well-characterized at the molecular level, how these pathways converge to reshape the tumour microenvironment (TME) and influence immunotherapy outcome remains unclear.

In this preprint (not peer reviewed), Serganova, Colombo, et al. demonstrates how glycolytic tuning impairs tumour growth through effects on the local vascular and immune landscape. In high glycolytic B16F10 melanoma and 4T1 triple-negative breast cancer models, knockdown of the  glycolytic enzyme LDHA not only reduced glycolytic capacity of tumour, but also normalized vasculature, increased lymphangiogenic factors, high endothelial venules (HEVs), and promoted tumour-specific CD8+ T cell egress to draining lymph nodes. Analysis of several human  cancers revealed that glycolytic signatures positively correlated with  features of neo-angiogenesis and inversely with HEV abundance and  cytolytic activity.  These findings reveal a mechanism by which tumour glycolytic capacity drives vascular abnormalities in the TME, ultimately contributing to poor outcomes.

To address whether normalizing the tumour vasculature could restore responses to immunotherapy, the authors  combined anti-angiogenesis therapy low-dose anti-VEGFR2 with  anti-CTLA-4. This combination reduced metastasis and extended survival  in glycolytic 4T1 and B16 tumours, observations that were associated with normalized vasculature and mobilization of central memory CD8+ T cells. Central memory CD8+ T cells express CD62L, which is a ligand bind to PNAd expressed on HEVs. Depleting CD8+ T cells or blocking PNAd HEVs abolished the therapeutic benefit. Strikingly, this combination therapy provided no benefit in LDHA-KD tumours and even appeared to compromise vascular normalization. In  hepatocellular carcinoma, where anti-VEGF is a frontline therapy  alongside anti-PD-L1, combination therapy significantly improved  survival of patients with tumours enriched for glycolytic signatures, but not in those with low rates of glycolysis.

This study supports the idea that immunotherapies targeting tumour angiogenesis and immune activation are most beneficial in the settings of dysregulated tumour vasculatures. It also highlights the importance of assessing tumour glycolytic state to identify patients most likely to benefit from combination therapy, while positioning metabolic reconditioning as  a strategy to broaden the efficacy and reach of cancer immunotherapy.

Limitations

Although  this study comprehensively investigated how tumor glycolytic state  shapes tumor vasculature and immunotherapy response using both mouse  models and human data, several limitations remain.  First, the authors only evaluated anti-CTLA4 therapy in combination with anti-angiogenic treatment, leaving it unclear whether similar  outcomes would be observed with other clinically relevant immune checkpoint inhibitors. Second, while the study elucidated mechanisms by which combination therapy  restores vascular normalization and enhances antitumor immunity in  highly glycolytic tumors, the mechanisms underlying its detrimental  effects on vascular normalization and immune cytolytic activity in  low-glycolytic tumors remain unclear. Whether these effects can be reversed through metabolic reprogramming also needs further investigation. Finally, although CD8+ T cell depletion and PNAd on intratumoral HEV blockade support the proposed mechanism, these approaches do not directly demonstrate the role of central memory CD8+ T cells. Additional experiments, such as selective depletion or adoptive transfer of central memory CD8+ T cells, would provide stronger evidence for the authors’ proposed mechanism.

Significance/Novelty

This study is the first to systematically establish a mechanistic link between cancer cell glycolytic capacity, and structural tumour vascular remodelling in vivo. It is also the first to investigate how tumours with different glycolytic states respond to combined immunotherapy and anti-angiogenic therapy. The findings propose tumour glycolytic state as a potential biomarker to stratify patients for combination therapy of immunotherapy and antiangiogenesis. A major strength of the study is the use of two distinct mouse tumour models and multiple complementary experimental approaches to support  the conclusions. Furthermore, validation of the key findings using human  datasets greatly enhances the translational relevance and overall impact of the work.

Credit

Reviewed by Giang Pham as part of a cross-institutional journal club between the Icahn School of Medicine at Mount Sinai, the University of Oxford, the Karolinska Institute and the University of Toronto.

The author declares no conflict of interests in relation to their involvement in the review.